The present invention relates to the field of electronic circuitry for the coincidence readout of solid-state detector signals, especially for application in the field of nuclear medical imaging.
In the field of nuclear medical imaging, solid state detectors are frequently used in order to directly detect the incidence of high energy photons, either those transmitted through the body from a high energy source, such as an X-ray source, or those emitted from radioactive isotopes injected or ingested into the body of a patient. In gamma-ray imaging applications, the detectors must typically be able to determine the energy of the gamma-ray photons emitted from the patient, and the position of incidence on the detector array.
In positron emission tomography (PET), also known as electronic collimation, isotopes that emit positrons are injected or ingested into the body of the examined patient. Each of the emitted positrons annihilates with an electron to produce a pair of 511 Kev photons propagating along the same line but in opposite directions and out of the patient""s body. The 511 Kev photons are detected by a camera which has two separate detector heads, which determine the position where the photons interact with the detector heads and the energy of these interacting photons using coincidence detection methods. Photons of the same pair are emitted simultaneously. Accordingly, the detection time of one photon of a pair should differ from the detection time of the second photon, only by a small time interval, xcex94t which depends, among other factors, on the time resolution of the system, and on the different time of flight of each photon to its corresponding detector head. The rate of the measured events in the detector heads determines the average time xcex94T between two followings events. Two photons are considered as being related to the same pair when they are detected by the two different detector heads of the camera within a time difference xcex94t, which satisfies the condition xcex94t less than xcex94T. The coincidence method is based on the detection of the time of the first impacting photon, and the use of that temporal information in deciding whether the following impacting photon is related to that event or not. If the criteria for coincidence are met, a coincidence trigger pulse is generated for informing the signal detection channels to process the signals accordingly.
Another form of gamma ray imaging is known as Single Photon Emission Computerized Tomography, or SPECT. In this method, lower energy photons are detected, such as the 140 keV photons emitted by the decay of Technecium-99 or Thallium-201 previously injected or ingested into the body. In this method, the photons emitted from the body of the patient are typically passed through a lead collimator, in order to ensure that only photons propagating in a straight line are used to produce the image, so that the image is a true representation of the source.
The detectors used in such imaging cameras are constructed of arrays of separate detection modules, each of which itself may have an array of several hundred separate detection areas, typically in the form of pixellated anodes. In a commonly used configuration, each module has 256 individual anodes, and each anode is connected to its own charge sensitive amplifier and signal processor, such that each is an effectively independent detector. Multiple detector readout channels from one module are often integrated into an ASIC (Application Specific Integrated Circuit). Each pixellated detector circuit is able to determine three pieces of data associated with the photons it detects:
(a) the point in time of the photon impact, known as the trigger time, this being important for the coincidence type of measurement;
(b) the energy of the impacting photon, determined from the amount of charge collected; and
(c) the position of impact, determined by means of an address which the detection channel of each anode pixel transmits together with the detection data it has collected.
A threshold level is used to discriminate between random noise in the detector, and a real incident photon. Since the detection channel associated with each detector pixel is only able to handle one detection event at a time, the detection circuitry must be programmed to reject any signals arriving substantially simultaneously in one ASIC from multiple photon impacts in the module. Substantially simultaneously is defined as being within the time taken for the detection circuitry to process and measure the arriving photon. In such a situation, the arrival of a second photon within the coincidence time, (generally of the order of up to tens of nanoseconds) of a first photon causes both signals to be rejected, in order to prevent corruption of the data of either of the photons. This process is known in the art as xe2x80x9cpile-up rejectionxe2x80x9d.
Because of the advisability of using low doses of radiation in the patient, the flux of imaging photons detected is very low. Furthermore, in SPECT imaging, the collimator typically transmits only 10xe2x88x924 of the incident flux, such that the detected flux is even further reduced. The importance of utilizing every piece of imaging information in the photon flux is thus of great importance, and every effort should be made not to lose any such information because of detection limitaitons.
As a result of the low flux levels used, simultaneous impact from the patient in a single module or ASIC, whether from direct emission or from Compton scattering within the patient""s body, is infrequent. In a single detector channel, this eventuality is even rarer. As a result, the process of pile-up rejection does not generally result in the loss of any significant data from source scattering or simultaneous emission.
However, there are two other processes which can result in the loss of imaging information from the incident flux of photons from the patient""s body. Firstly, Compton scattering can occur, not only in the patient""s body, but also in the detector bulk itself. This is a much more common phenomenon, and much more serious, as it can affect an appreciable percentage of all photon detection events. When this occurs, the incoming photon makes an initial impact within the detector, and gives up part of its energy in producing a charge of excited electron-hole pairs. The electrons in this cloud of charge then drift towards the anode under the influence of the field present in the detector, and appear on the anode opposite the point of first impact. The secondary scattered photon continues its path within the detector, at a lower energy, until it makes a second impact within the detector bulk, again resulting in another cloud of carriers, the electrons of which are collected at the closest anode. This anode may be in a pixel a considerable distance from the pixel of the initial impact, especially for high energy photons. (The hole motion has been neglected for the purposes of this explanation). As a result of this process, the primary and secondary impacts are detected by different pixels at different times in the detector module. Furthermore, they each have a different energy, as the energy of the incident photon is shared between the two impacts, and neither therefore has the expected energy of the emitted photon.
The second process which can result in the loss of imaging information, occurs when a photon impacts the detector very close to the border between two pixels. This is known as sharing. When this happens, the cloud of charge carriers is approximately divided between two neighboring pixels, and some of the electrons are collected by one anode, and others by the neighboring anode. The incident photon is thus measured as if it were two separate impacts, in neighboring pixels, neither of which has the characteristic photon energy sought for constructing the image, and this event too would be rejected from the imaging process.
Sharing can be even more complex if the impact occurs near the junction of four pixels, in which case the photon energy detected is shared between four anodes. Furthermore, sharing can occur in combination with inter-detector Compton scattering, thus further complicating the issue of relating the charge signal measured by each pixel with the energy and location of an incoming photon.
A result of either of the above processes or of their combination, is that a single incident photon of the correct energy for use in the image reconstruction, may appear to be detected as a number of photons of lower energy, impacting the detector at a number of different locations, and at different times. It would be important to be able to collect all of the impact data, and from the location and energy of the events detected, to determine whether the original photon incidence is useful for constructing the image. In order to do this, it is necessary to be able to provide accurate readouts of a number of apparently multiple events occurring at very close intervals.
In prior art methods, in order to avoid confusion between different detected events, the first apparent photon detected over the threshold level is fully measured, i.e. its time of arrival and its location are recorded, and its energy measured. Immediately after the impact, the signal detection circuits in that module are blocked, to avoid corrupting the data being measured by another photon impact, which in the situation described above, could be a secondary impact of the same photon, or a result of charge sharing, or both. The required blocking time for this process is the time taken to process the data arising from an impact, and is dependent on the electronic circuits used, but typically ranges from less than 1 microsecond to over 10 microseconds.
As a result of the blocking process, the second or later apparent impacts are rejected, since the blocking time may well be longer than the arrival time of secondary or shared charges. Since, however, because of dosing considerations, the incident flux from the patient is low, it would be very advantageous not to lose any imaging information carried by either of the photons, as mentioned above. Therefore, methods have been derived in the prior art whereby both of the apparent scattered photons are detected and used in generating the image required.
One commonly used prior art method relies on a process of freezing all of the information accumulated a certain time after an impact is detected, and then serially scanning the charges accumulated by all of the pixels in the vicinity of the impact pixel, and sending the information for processing. This method has a number of disadvantages. Firstly it is slow because of the serial method of scanning, and because of the large number of pixels that have to be scanned. For low energy photons, it is usually sufficient to scan only the eight nearest neighbors of the impact pixel, but for high energy photons, the scattered photons may reach a much wider area. Secondly, it is electronically complicated to perform, and is noisy.
Another method of reading all of the scattered photon information from a single event is described in U.S. Pat. No. 5,656,818, to one of the present inventors, hereby incorporated by reference in its entirety. This method uses a single channel data transfer method, whereby all of the channels are commonly controlled with a single reset. This method thus has the disadvantage that it is comparatively slow, that it has no multi-channel readout ability, and that it has no coincidence measurement facility.
There therefore exists a serious need for a method and apparatus for reading out the charge signals detected by a solid state detector, in such a manner that even closely occurring multiple events such as temporally closely detected photons arising from scattering within the detector bulk, are all read in a speedy and efficient manner, without the loss of imaging information because of any significant dead time.
The disclosures of all publications mentioned in this section and in the other sections of the specification are hereby incorporated by reference, each in its entirety.
The present invention seeks to provide a new electronic signal processing system that can detect, amplify, noise-filter and transfer electronically to an external Data Acquisition System (DAQ), the position, signal amplitude and timing of charge signals arising as a result of impinging photons or charged particles on a solid state radiation detector array consisting of a plurality of pixellated sensor elements. The system is coincidence-enabled, and also allows more than just one detector element of a detection array module to be read-out simultaneously, and at a high rate. The other detector elements of interest to be read out may preferably be the elements immediately surrounding the primary element, such as results from event sharing, or elements significantly further away such as those caused by high energy Compton scattering within the detector.
There is thus provided in accordance with a preferred embodiment of the present invention, a system consisting of a plurality of individual detection channels that coexist with other identical parallel detection channels, arranged in a multichannel structure, in which all of these independent detector channels share a common readout architecture. Such a multichannel structure is called a Chip-on-board (COB) unit, and the number of parallel channels in a single COB unit is typically a few hundred. Physically, each COB unit is typically integrated in a single ASIC. According to this preferred embodiment of the present invention, all of these independent detector channels are kept open in a ready state to receive the input of a detected event. When such an event is detected and read-out in any single channel, only the channel reading the data is blocked from receiving another set of data. This blocking is preferably performed by a self-generated process. All of the other channels are still ready to receive data from a different detected event. When the active channel has completed processing and transferring its received data, it resets itself without any external interference or command. If the event detected by that channel was not a coincidence event, then the channel resets itself after a predetermined internal decay time of the data.
A further advantage arising from the above-mentioned properties, is that the system can operate at a higher rate than prior art detection systems, since all of its detection channels, other than those currently measuring an event, are always open and waiting to receive a new event, without the need for any comparatively time-consuming common reset signal.
In a complete gamma-camera, there are generally many identical COB units. The system of the present invention may be advantageously used for processing and reading out the signals from a gamma camera, such as that described in the co-pending Israel Patent Application for a Coincidence Gamma Ray Detector, some of whose inventors are inventors of the present application.
It should be pointed out, though, that this invention is not limited to Nuclear Medicine and gamma-ray detection. The electronic architecture can advantageously be used also with other types of sensors and types of radiation other than gamma-rays, and for detecting multiple impacts of charged particles.
In accordance with yet another preferred embodiment of the present invention, there is provided a signal readout system for a solid state detector array consisting of a plurality of detection channels, wherein the decision to output a signal detected by a channel is determined by the content of that channel.
There is further provided in accordance with yet other preferred embodiments of the present invention, a signal readout system as described above, wherein the decision also depends on the presence of an external signal, which could be a coincidence signal derived from the outputs of at least two heads of a coincidence gamma camera, or a fixed signal.
In accordance with still another preferred embodiment of the present invention, there is provided a signal readout system for a solid state detector array, consisting of a plurality of detection channels, and a switching network, wherein the switching network outputs data simultaneously from at least two of the plurality of detection channels. That data could be output in parallel.
There is further provided in accordance with still another preferred embodiment of the present invention, a signal readout system as described above, and wherein those of the plurality of detection channels which are not outputting data are in an effectively continuous state of readiness to receive data from a further detected signal.
In accordance with a further preferred embodiment of the present invention, each of the plurality of detection channels outputs data independently of the status of the other ones of the plurality of detection channels.
There is also provided in accordance with a further preferred embodiment of the present invention, a signal readout system as described above, and wherein the switching network consists of an arbitrator and a plurality of multiplexer units, the arbitrator directing the outputs of at least two of the plurality of detection channels into at least two of the plurality of multiplexer units according to the content of the at least two detection channels.
In accordance with yet another preferred embodiment of the present invention, there is provided a signal detection channel for use in a readout system for a solid state detector array, consisting of a peak-and-hold circuit operative for capturing the signal, and wherein the peak-and-hold circuit is made operative according to a trigger signal generated according to the content of the signal detection channel itself.
Furthermore, in accordance with yet more preferred embodiments of the present invention, the trigger signal is generated according to the content of the signal detection channel itself only when an external signal is present, and that external signal may be a coincidence signal derived from the outputs of at least two heads of a coincidence gamma camera, or a fixed signal.
In accordance with yet another preferred embodiment of the present invention, there is provided a method of reading out signals from a solid state detector array having detector elements, consisting of the steps of providing a plurality of detection channels, essentially one for each element of the array, and outputting a signal detected by a channel according to the content of that channel.
There is further provided in accordance with yet another preferred embodiment of the present invention, a method of reading out signals from a solid state detector array having detector elements, consisting of the steps of providing a plurality of detection channels, essentially one for each element of the array, and a switching network, and causing the switching network to output data simultaneously from at least two of the plurality of detection channels.
In accordance with still another preferred embodiment of the present invention, there is provided a method of reading out signals from a signal detection channel of a solid state detector array, consisting of the steps of providing the signal detection channel with a peak-and-hold circuit, and capturing the signal using the peak-and-hold circuit according to a trigger signal generated according to the content of the signal detection channel itself.